Synthesis and activity of fibrillogenesis peptide inhibitors related to the 17-21 β-amyloid sequence

Article abstract of DOI:10.1016/j.ejmech.2008.03.036

Peptide derivatives 1-5, incorporating synthetic non-proteinogenic amino acids, related to the β-amyloid 17-21 fragment of the amyloidogenic Aβ_1-40, and the N-protected decapeptide 6, corresponding to a dimeric sequence of the same fragment, ha

Full text of DOI:10.1016/j.ejmech.2008.03.036

Available online at www.sciencedirect.com
European Journal of Medicinal Chemistry 44 (2009) 179e189
http://www.elsevier.com/locate/ejmech
Original article
Synthesis and activity of fibrillogenesis peptide inhibitors
related to the 17e21 b-amyloid sequence
b
b
b
Cesare Giordano a, , Annalisa Masi , Aldo Pizzini , Anna Sansone ,
*
Valerio Consalvi c, Roberta Chiaraluce c, Gino Lucente b
a Istituto di Chimica Biomolecolare del CNR, Universita` ‘‘La Sapienza’’, P.le A. Moro 5, 00185 Roma, Italy
b Dipartimento di Studi Farmaceutici, Universita` ‘‘La Sapienza’’, P.le A. Moro 5, 00185 Roma, Italy
c Dipartimento di Scienze Biochimiche, Universita` ‘‘La Sapienza’’, P.le A. Moro 5, 00185 Roma, Italy
Received 31 October 2007; received in revised form 21 February 2008; accepted 20 March 2008
Available online 8 April 2008
Abstract
Peptide derivatives 1e5, incorporating synthetic non-proteinogenic amino acids, related to the b-amyloid 17e21 fragment of the amyloido-
genic Ab1e40, and the N-protected decapeptide 6, corresponding to a dimeric sequence of the same fragment, have been synthesized. These
compounds were designed by using Soto’s pentapeptide AceLPFFDeNH2 (iAb5p) as lead compound. Their activity as inhibitors of fibrillo-
genesis and stability against enzymatic degradation have been determined. Compounds 1, 5 and 6 are potent inhibitors in comparison to the lead
compound. Exposure to chymotrypsin of peptide derivatives 1e5, all containing unnatural amino acids, shows increased stability as compared
with iAb5p and 6. Conformational properties of the new compounds have been determined by CD and FT-IR spectroscopies.
Ó 2008 Elsevier Masson SAS. All rights reserved.
Keywords: Alzheimer’s disease; b-Amyloid; Fibrillogenesis; Aggregation inhibitor; Fluorescence assay; Scanning electron microscopy
1. Introduction
amyloid fibril formation and deposit are due to conformational
switching from a-helical to b-sheet structure during amyloido-
genesis [1]. On this basis, a rational pharmacological approach
for prevention of amyloid formation and deposit would be to
find compounds able to specifically stabilize the a-helical
conformations [3] and/or destabilize the b-sheet conformations
of the amyloidogenic protein [4,5]. Several recent papers deal
with this topic trying to gain insight into the mechanism of
amyloid fibril formation and the role played by different factors
such as H-bonding, peptide backbone chirality and composition
[6e12].
The design of a therapeutic strategy against this pathology
initiated from the observation that short synthetic peptides are
capable to interact with Ab without becoming part of b-sheet
structure. These synthetic peptides, named b-sheet breaker
peptides (BSBp), may destabilize the amyloidogenic Ab
conformer and hence preclude amyloid formation [13].
Since fibril formation could be impaired by substitution of
hydrophilic for hydrophobic residues in the hydrophobic re-
gion of the Ab1e40 sequence [14], we chose as model of
Amyloidoses are a class of diseases in which soluble proteins
or peptides accumulate in the form of amyloid aggregates which
are deposited as plaques in the extracellular space. Plaques are
commonly observed in diseases of completely unrelated origin,
such as Parkinson’s disease, type II diabetes, prion disorders
and Alzheimer’s disease (AD). AD is the most common form
of dementia in adults and the neuritic plaques in the brain con-
tain extracellular deposit of amyloid b-protein (Ab) [1,2]. Ab is
composed of a 39e43 amino acid long peptide produced by
cleavage from a larger amyloid precursor protein (APP). The
Abbreviations: DMF, dimethylformamide; EDC, N-(3-dimethylamino-
propyl)-N0-ethylcarbodiimide hydrochloride; EDT, ethane dithiol; HOBT, 1-
hydroxybenzotriazole; PBS, phosphate buffer saline; SEM, scanning electron
microscopy; TEA, triethylamine; TFA, trifluoroacetic acid; TPS,
triisopropylsilane.
* Corresponding author. Tel.: þ39 06 49913237; fax: þ39 06 491491.
E-mail address: cesare.giordano@uniroma1.it (C. Giordano).
0223-5234/$ - see front matter Ó 2008 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2008.03.036

180
C. Giordano et al. / European Journal of Medicinal Chemistry 44 (2009) 179e189
inhibitor the synthetic five-residue b-sheet-breaker peptide de-
rivative AceLeu-Pro-Phe-Phe-Asp(OH)eNH2 (AceLPFFDe
NH2; iAb5p). This compound, previously reported by Soto
and coworkers [15], in addition to N- and C-terminal protec-
tions has a proline and an aspartic residue which substitute va-
line and alanine, respectively, in the 17e21 (LVFFA) fragment
of the 1e40 sequence and possesses a series of promising
properties: it has been shown to cross the bloodebrain barrier,
to inhibit (and disassemble) the formation of amyloid fibrils in
vitro and in vivo, to prevent Ab neurotoxicity in cell culture, to
arrest amyloid deposition and to induce dissolution of pre-
formed plaques in rat brain model of amyloidosis [5,16,17].
Particularly effective in determining the iAb5p behaviour is
the incorporation of the proline residue, a strategy commonly
used to induce secondary structures and inhibit b-sheet forma-
tion. It is worth noting in this context that incorporation of pro-
Compounds 1e4 maintain the original sequence and pro-
tecting groups of iAb5p with the exception of substitutions
at Pro residue. The four-membered azetidine skeleton was in-
corporated in analogues 1 and 2; in particular (see Scheme 1)
compound 1 contains the residue of (S )-azetidine-2-carboxylic
acid (Aze) while in 2 the (2R,3R)-trans-3-phenyl substituted
analogue (PhAze) was incorporated. As compared to Pro, pep-
tides containing the smaller ring of Aze show comparable be-
haviour toward the cisetrans isomerism but higher backbone
flexibility due to the lower steric interference exerted by the
ring atoms [19]. The more constrained ring of the Aze should,
however, maintain and probably accentuate some of the ex-
pected consequences of the Pro residue incorporation, such
as alteration of the backbone conformation and incompatibil-
ity with the peptide bond geometries found in the b-sheet
bonding network. In analogues 3 and 4 the Pro residue was
homologated to give a b-amino acid structure. Compound 3
contains (S )b3-HPro [20] while in analogue 4 the Pro homol-
ogation has been combined with the replacement of the
carboxyl with a sulfonyl group, thus generating a sulfonamide
junction and an a-peptide/b-peptidosulfonamide hybrid
[21,22]. As it is well documented, the incorporation of b-
aminocarboxylic or b-aminosulfonic acid residues into natural
peptides has profound effect on both backbone conformation
and metabolic stability. In particular, sulfonamido-b-peptides
and related a/b hybrids represent a very attractive class of
compounds due to the peculiar chemical and structural proper-
ties of the SO2eN(R) junction. This bond, which is com-
pletely stable to proteases and possesses two negatively
charged oxygen atoms directly bound to the tetrahedral sulfur,
can significantly alter polarity and H-bonding patterns of the
native counterparts. The approach based on the incorporation
of a CH2eSO2eNH junction, replacing the usual COeNH
peptide bond, has also been applied in the case of compound
5. This latter compound resembles the iAb5p analogue
(MeSareLPFFDeNH2) previously studied by Soto’s group
[15] and contains, as N-terminal residue, N,N-dimethyl-taurine
(N,N-dimethy-2-aminoethanesulfonic acid; Me2Tau) residue
in place of N,N-dimethylglycine (Me2Gly or MeSar). Finally,
in a different approach compound 6 has been synthesized. This
is the iAb5p dimer in which the AceLPFFDeOH unit has
been joined to the N-terminal group of the pentapeptide
LPFFDeNH2 giving rise to the N-protected decapeptide
derivative 6.
line in two series of Ab fragments (Ab15e23 and Ab12e26
)
showed that residues 17e23 make up the b-sheet core of these
fragments [18].
The iAb5p molecule has been adopted by Soto’s group as
lead compound and several strategic chemical modifications
have been applied in order to improve its activity and meta-
bolic stability [15]. Ca-Carbon and N-amide bond methylation
showed that the major cleavage site by peptidases is at the
ProePhe bond. Furthermore, substitution of these two amino
acids with unnatural residues was found to be highly detrimen-
tal for in vitro activity in all examined derivatives. Replace-
ment of Pro with cysteine-derived pseudoproline residues,
able to stabilize cis conformation around the COeN bond,
decreased the activity, thus suggesting that trans configurated
COePro amide bond is required for the activity. Furthermore,
replacement of the N-terminal acetyl group with residues of N-
methylglycine (sarcosine, Sar) or N,N-dimethylglycine (N-
methylsarcosine, MeSar), leading to SareLPFFDeNH2 and
MeSareLPFFDeNH2, resulted in a decrease of peptidase
recognition and high activity against fibrillogenesis [15].
The design of the pseudopeptides in this report aimed to
disrupt the hydrogen bonding that stabilizes the b-structure
of fibrillar amyloid aggregates. Furthermore, taking into ac-
count the short half-life of peptides in circulation, we focused
also on the possibility to obtain more stable sequences taking
into account the position-2 sensitivity of iAb5p toward prote-
olysis. According to the above considerations, we synthesized
the analogues 1e6 of the iAb5p reported in Table 1.
Table 1
Sequence and mass spectra analysis of iAb5p and its synthetic analogues
Peptide
Sequencea
M (Naþ) obsd (MW calcd)
b
iAb5p
AceLeu-Pro-Phe-Phe-Asp(OH)eNH2
AceLeu-Aze-Phe-Phe-Asp(OH)eNH2
701.5302 (678.3533)
687.1848 (664.3220)
1
2
3
4
5
6
a
AceLeu-(2R,3R)-PhAze-Phe-Phe-Asp(OH)eNH2
AceLeu-b-HPro-Phe-Phe-Asp(OH)eNH2
AceLeu-Pro-j[CH2SO2]-Phe-Phe-Asp(OH)eNH2
Me2Tau-Leu-Pro-Phe-Phe-Asp(OH)eNH2
AceLeu-Pro-Phe-Phe-Asp(OH)-Leu-Pro-Phe-Phe-Asp(OH)eNH2
763.3305 (740.3533)
715.2387 (692.3533)
751.1428 (728.3203)
794.0246 (771.3625)
1321.5274 (1297.6383)
Modifications are shown in bold letters.
See Ref. [15].
b

C. Giordano et al. / European Journal of Medicinal Chemistry 44 (2009) 179e189
181
P-AA-X + H-Phe-OMe
Peptide
P
X
AA
i
Ph
O
1
OH
N
CO
Ph
O
P-AA-Phe-OMe
ii
OH
2
3
4
Ph
N
CO
t
P-AA-Phe-Phe-Asp(O Bu)-NH2
Ph
O
N
OH
Cl
iii
CO
O
O
t
Ac-Leu-AA-Phe-Phe-Asp(O Bu)-NH2
Ph
O
N
iv
SO2
(1-4)
Ac-Leu-AA-Phe-Phe-Asp(OH)-NH2
Scheme 1. Synthesis of pseudopeptides 1e4. Reagents: (i) for 1e3: EDC, HOBt, TEA, DCM; for 4: TEA, DCM; (ii) a: 2 N NaOH, MeOH; b: HePhe-Asp(OtBu)e
NH2, EDC, HOBt, TEA, DCM; (iii) a: for 1, 3 and 4: H2/PdeC, TFA, MeOH; for 2: 4% formic acid in MeOH, Pd black; b: AceLeueOH, EDC, HOBt, TEA,
DCM; (iv) TFA.
2. Results
For the synthesis of peptide 5 (Scheme 2), CbzeTau-Leue
OH was coupled with Pro-PheeOMe. The tetrapeptide was
then subjected to alkaline hydrolysis and coupled with Phe-
Asp(OtBu)eNH2. After N-deprotection by catalytic hydroge-
nation, the taurine-containing peptide was N,N-dimethylated
by reductive alkylation with 35% aqueous formaldehyde and
sodium cyanoborohydride [23].
Treatment with TFA afforded the required final free pep-
tides. All peptides were purified by HPLC and characterized
by mass spectrometry (see Table 1). (2R,3R)-(1-Benzyl-3-phe-
nyl-azetidine-2-yl)-carboxylic acid hydrochloride, required
for the synthesis of the peptide 2, was prepared starting
from (R)-phenylglycinol according to a published procedure
[24]. The Cbze(S )-b-homoproline-derived sulfonylchloride
(CbzeProej[CH2SO2]eCl) 3, required for the preparation
of the peptide 4, was synthesized according to Scheme 3 start-
ing from the corresponding sulfonic acid 1 [25]. After N-
protection with benzyl-chloroformate, the sodium salt 2 was
converted into the corresponding sulfonylchloride 3 by treat-
ment with phosgene [26].
2.1. Chemistry
Preparation of peptides 1e5 was performed by conven-
tional methods in solution, using the carbodiimide method
for coupling steps, except for the sulfonamido derivatives
which were obtained by using sulfonylchlorides. For the syn-
thesis of the pseudopeptides 1e4 (Scheme 1), the N-protected
pseudoaminoacid (PeAAeX) was coupled with PheeOMe
and the resulting dipeptide methylester was subjected to alka-
line hydrolysis. Successive coupling with Phe-Asp(OtBu)e
NH2 gave the tetrapeptide which was N-deprotected and
coupled with AceLeueOH.
Cbz-Tau-Leu-OH + H-Pro-Phe-OMe
i
The peptide iAb5p and its analogue 6 were synthesized by
Fmoc solid-phase synthetic protocols on Rink-amide resin
according to standard protocols [27]. The crude peptides
were purified by RP-HPLC using a Bondapac C-18 column
(1.9 ꢀ 30 cm) with elution at 8 mL/min by a linear gradient
of 10e60% acetonitrile in 0.1% aqueous trifluoroacetic acid
in 30 min.
Cbz-Tau-Leu-Pro-Phe-OMe
ii
t
Cbz-Tau-Leu-Pro-Phe-Phe-Asp(O Bu)-NH2
iii
t
N,N-Dimethyl-Tau-Leu-Pro-Phe-Phe-Asp(O Bu)-NH2
iv
SO3H
i
ii
SO3Na
SO2Cl
(5)
N,N-Dimethyl-Tau-Leu-Pro-Phe-Phe-Asp(OH)-NH2
N
H
N
N
1
2
3
Scheme 2. Synthesis of pseudopeptide 5. Reagents: (i) EDC, HOBt, TEA,
DCM; (ii) a: 2 N NaOH, MeOH; b: HePhe-Asp(OtBu)eNH2, EDC, HOBt,
TEA, DCM; (iii) a: H2/PdeC, TFA, MeOH; b: (CH2O)n/NaBH4CN, MeCN;
(iv) TFA.
Cbz
Cbz
Scheme 3. Synthesis of derivative 3. Reagents: (i) CbzeCl, 1 N NaOH; (ii)
COCl2/toluene.

182
C. Giordano et al. / European Journal of Medicinal Chemistry 44 (2009) 179e189
2.2. In vitro activity and stability
The in vitro inhibitory activity of the new analogues 1e6
toward amyloid fibril formation has been determined by an
amyloid-specific dye binding analysis and compared with
that of a sample of the reference model iAb5p under the
same conditions (Fig. 1). The quantitative evaluation of the
effect of compounds 1e6 on in vitro fibrillogenesis was per-
formed with thioflavin T (ThT), a fluorescent dye, which as-
sociates with amyloid fibrils with a concomitant fluorescence
enhancement in response to the fibril aggregation grade [28].
The preparation of Ab1e40 according to Fezoui et al. [29]
allowed us to easily obtain fibril formation reactive to ThT
on a 24-h time scale. The first experimental data point was
taken after 30 min incubation of Ab1e40 in the absence or
presence of inhibitors (Fig. 1A). In order to decrease the
experimental error due to sample evaporation, the incubation
was carried out at 20 ꢁC for a maximum time of 24 h. The
results are shown in Fig. 1A and compared with the fibril
formation in the absence of inhibitor and in the presence of
iAb5p under the same conditions. A bar plot (Fig. 1B) of
the amount of fibril formation after 24 h incubation of the
Ab1e40 fraction in the presence of the inhibitors indicates
that compounds 1, 5 and 6 show activity 1.2-, 1.5- and 2.2-
fold higher than iAb5p, respectively, while compounds 2e4
do not show any significant inhibition of Ab1e40 fibril forma-
tion (data not shown).
250
200
150
100
50
A
0
0
5
10
15
20
25
Time (h)
250
B
The effect of inhibitors on Ab1e40 fibrillogenesis was stud-
ied also by electron microscopy. Fig. 2 shows SEM images of
Ab1e40 after incubation at 20 ꢁC in the absence (Fig. 2A and
B) and presence (Fig. 2C and D) of compound 5. In the ab-
sence of the inhibitor, Ab1e40 sample shows the formation
of protofibrils less than 1 mm in length [30,31] which become
more clearly visible after 24 h incubation (Fig. 2B). In the
presence of the inhibitor (Fig. 2C and D), the protofibrils as
well as any kind of fibrillar structure are absent; the few glob-
ular structures visible do not show any morphological resem-
blance of amyloid aggregates and are presumably formed by
the inhibitor, as suggested by SEM images of the inhibitor
alone (data not shown).
In agreement with the experimental hypothesis, the mod-
ification of the residues more exposed to the proteolytic
cleavage increases the resistance of the compound to prote-
olysis (Fig. 3). The iAb5p analogues 1e5, all containing un-
natural amino acids, after exposure to chymotrypsin show,
under the adopted experimental conditions, increased stabil-
ity in comparison with the parent peptide iAb5p and to its
decapeptide analogue 6. After 2 h incubation at 37 ꢁC,
only 12 and 22% of iAb5p and peptide 6, respectively,
were detected, while 35e45% of the pseudopeptides 1e4
and 70% of the pseudopeptide 5 survived. After 20 h incuba-
tion, the iAb5p and the peptide 6 were near totally degraded,
while 15e20% of analogues 1e4 and 55% of analogue 5 re-
mained intact. The analogue 6 together with the unmodified
iAb5p model, both containing only proteinogenic residues,
shows the expected lowest stability toward enzymatic
degradation.
200
150
100
50
0
iAβ5p
1
iA 5p derivatives
5
6
Fig. 1. In vitro activity of iAb5p derivatives 1e6. All compounds were tested
for activity using the ThT fluorescent assay for quantification of amyloid fibril
formation by Ab1e40. (A) Ab1e40 fraction (about 100 mg) was incubated in
100 mL of 30 mM Tris/HCl buffer at pH 7.4 at 20 ꢁC with and without
10 M excess of the synthesized compound (> control, B 1, - 2, , 3,
4,
bition of amyloid formation and are expressed relative to the activity of iAb5p
5, ꢂ 6,
iAb5p). (B) Values correspond to the percentage of inhi-
A
:
6
(100%).

C. Giordano et al. / European Journal of Medicinal Chemistry 44 (2009) 179e189
183
Fig. 2. Effect on Ab1e40 aggregation. SEM images of Ab1e40 (100 mg/100 mL of 30 mM Tris/HCl, pH 7.4) incubated at 20 ꢁC in the absence (A and B) and pres-
ence (C and D) of 10 M excess of compound 5. Aliquots of 5 mL were withdrawn after 30 min (A and C) and 24 h (B and D), frozen, lyophilized and gold coated,
by sputtering, prior to SEM analyses.
120
2.3. Spectroscopic characterization in solution
3
iAβ5p
1
4
5
The peptides have been analyzed by far-UV CD (180e
250 nm) (Fig. 4) and by FT-IR in H2O and in D2O, respec-
2
100
6
tively (Fig. 5). Far-UV CD spectra of peptides iAb5p, 2, 3
and 5 show a weak positive band (nep*) at around 220 nm
80
and a strong negative band around 200 nm (pep*), less pro-
nounced for peptide 2, which can be assigned to polyPro II
conformation [32,33]. The iAb5p and 6 show a weak negative
60
band around 235 nm which can resemble that of b-turn type II
[33] and could be attributed to the aromatic contribution to the
far-UV CD spectrum [34e36]. Similar redshifted negative (ne
p*) bands have been also reported for cyclic peptide and
attributed to a composite of class spectrum of II b-turn and
the g-turn [33]. Interestingly, the negative band at 230 nm,
present in iAb5p and in 6, is absent in spectra of 2, 3 and 5.
The far-UV CD spectra of 1 and 4, differently from the other
peptides, both show negative ellipticity at around 220 nm;
a zero intercept at around 190 nm is detected only for 4.
The far-UV CD spectra indicate that 2, 3 and 5 present
dichroic activity bands similar to those of the iAb5p leading
structure (Fig. 4). On the other hand, the ellipticity observed
for 1 and 4 indicates that the changes in the primary structure
have caused significant changes in the secondary structure
elements with respect to iAb5p. Peptides have been also ana-
lyzed by FT-IR spectroscopy, in D2O, in the amide I0, amide II0
and in the spectral region where the side chains of amino acid
40
20
0
0
5
10
15
20
25
time (h)
Fig. 3. Patterns of proteolysis of compounds 1e6 and iAb5p by chymotrypsin.
Aliquots of iAb5p and compounds 1e6 at a concentration of 0.76 mg/mL
were incubated in PBS with 0.1 mg/mL of chymotrypsin at 25 ꢁC. At the de-
sired time, aliquots were withdrawn, added to the same volume of the column
elution mixture to stop the reaction and injected on an RP-HPLC column Vy-
dac C18. The area under the peak (UV absorbance at 215 nm) corresponding to
the intact peptide was measured and compared with an equivalent sample in-
cubated in PBS.

184
C. Giordano et al. / European Journal of Medicinal Chemistry 44 (2009) 179e189
4000
0
A
0.06
6
0.05
1750
1500
2
1
1
5
-4000
-8000
-12000
-16000
-20000
0.04
4
3
0.03
6
0.02
0.01
0.00
1Aβ5p
1
2
3
4
5
6
1750 1700 1650 1600 1550 1500 1450 1400
iAβ5p
200
190
210
220
Wavelength (nm)
230
240
250
0
Fig. 4. Far-UV CD spectra of compounds 1e6 and iAb5p. All the spectra were
recorded at 20 ꢁC in ultra high quality water.
1750 1700 1650 1600 1550 1500 1450 1400
Wavenumber (cm-1)
residues are known to make contributions [37,38] (Fig. 5 and
Table 2).
B
FT-IR absorbance spectrum of iAb5p shows a maximum
centred at 1647 cmꢃ1 and a band at 1456 cmꢃ1, corresponding
to the amide I0 and amide II0 region, respectively. The amide I0
region presents two significant shoulders at 1671 and
1620 cmꢃ1. In the amide I0 region, the main peak at
1647 cmꢃ1 can be assigned to unordered loops [37,39] and
the 1671 and 1620 cmꢃ1 shoulders to b-turns [37,39] and poly-
Pro II [40], respectively. The region of antisymmetric stretch-
ing vibration nas COOꢃ mode of the unprotonated carboxyl
groups (1595e1550 cmꢃ1) and the region 1430e1390 cmꢃ1
corresponding to the symmetric stretching vibration ns COOꢃ
mode of the unprotonated carboxyl groups show the typical
contributions of the Asp residues [38,39]. The contribution of
Phe at 1498 cmꢃ1, Pro at 1470 cmꢃ1 and of das (CH3) of Leu
at 1440 cmꢃ1 is also observed [38,39]. Peptide 6 shows the
same bands observed for iAb5p, except for a decrease in inten-
sity of the shoulder at 1618 cmꢃ1. FT-IR spectra of peptides 1
and 2 show both a significant decrease in the 1672 cmꢃ1 and an
increase in the 1618 cmꢃ1 bands and for peptide 2 the main aꢃm1-
0.06
0.05
0.04
0.03
0.02
0.01
0.00
1750
1500
1750 1700 1650 1600 1550 1500 1450 1400
0
ide I0 band is broadened and shifted from 1647 to 1640 cm
.
The presence of the 1618 cmꢃ1 band in 1 is not paralleled by
the typical far-UV CD of polyPro II.
1750 1700 1650 1600 1550 1500 1450 1400
Wavenumber (cm-1)
The contributions of amide I0 region in the spectrum of pep-
tide 3 are comparable to those of iAb5p, except for a decrease
in the intensity of the 1618 cmꢃ1 band (Fig. 5 and Table 2).
Peptides 4 and 5 are characterized by the absence of the con-
tribution of the 1618 cmꢃ1 band and by a significant increase
in the intensity of the 1672 cmꢃ1 band. In the amide II0 region
the spectra of all the peptides are comparable and do not show
any significant difference. Notably, FT-IR spectra of all the
compounds show a decreased content of the unordered loop
Fig. 5. FT-IR spectra of compounds 1 and 6. Solution FT-IR spectra were mea-
sured at 20 mg/mL compound in 2H2O in a CaF2 cell with a 0.015 mm spacer
at 20 ꢁC. A total of 512 interferograms at 2 cmꢃ1 resolution were collected for
each spectrum. Absorbance spectra of compounds 1 (A) and 6 (B) amides I0
and II0 in 2H2O. Deconvolution of the spectra was used to identify the individ-
ual components in the amide I0 and amide II0 regions by curve fitting of the raw
spectra. The panels show the second derivative of the raw spectra. Insets show
the difference between the fitted curve and the original expanded 20 times.

C. Giordano et al. / European Journal of Medicinal Chemistry 44 (2009) 179e189
185
Table 2
Peak position and secondary structure assignment of the amide I0 of compounds 1e6 and iAb5p
Peak position (cmꢃ1
)
Area (%)
Assignment
iAb5p
1
2
3
4
5
6
1672
1647
1618
1581
1498
1465
1456
1440
1416
1398
4.45
45.9
4.39
18.28
0.78
14.78
1.61
5.81
1.73
3.09
5.08
1.42
10.01
27.10
11.00
14.63
1.15
6.14
29.76
0.04
14.75
26.23
1.44
5.79
b-Turn
30.20
10.02
18.61
0.21
25.08
15.27
12.44
0.18
39.90
3.20
22.03
3.08
10.62
2.63
7.76
1.48
3.65
Unordered loops
PolyPro II þ Pro
Aspa
11.14
4.03
12.40
1.13
8.09
Phe
22.39
0.87
6.51
23.67
0.70
8.70
10.10
2.47
17.38
3.36
Pro, Leub
Amide II0
Pro, Leuc
Aspd
11.58
12.50
5.74
13.30
4.48
5.95
14.59
7.07
6.48
2.60
3.51
4.05
8.50
6.15
Aspd
a
nas(COOH).
d(CH2).
b
c
das(CH3).
ns(COOH).
d
band, as indicated by the decrease in the area at 1647 cmꢃ1
,
The consequences on the inhibitory activity of structural
thus suggesting a decreased structural flexibility for all the
peptides. The relative content of the b-turn band is decreased
in 2 and increased in 3 and 5. Interestingly, the polyPro II band
at 1618 cmꢃ1, which is increased in 1e3, is absent in 4 and
decreased in 5, compounds both including a peptidosulfona-
mide moiety.
modifications performed at the N-terminal protecting group
of iAb5p have been previously examined by Soto’s group
[15]. Although the results reveal a complex picture, an inter-
esting indication is that iAb5p analogues possessing sarcosine
or N-methyl-sarcosine bound at the Leu amino group show the
highest activity. On the basis of these findings we synthesized
the analogue 5 which maintains the N-terminal tertiary amino
group as part of the b-amino-ethanesulfonic fragment bound to
the N-terminal Leu residue. In accordance with the previous
results [15] the sulfonamide-pseudopeptide 5 shows the high-
est activity among the here examined pentapeptide derivatives
(Fig. 1) and the highest enzymatic stability (Fig. 3). It can be
noted here that a significant resistance to the hydrolytic degra-
dation is found also in the case of compound 4 which shares
with 5 the presence of the sulfonamide junction in the
molecule.
3. Discussion
The formation of insoluble Ab deposit in the brain is the
main pathological trait of Alzheimer’s disease. The inhibition
of Ab fibril deposition might slow down or prevent the dis-
ease, according to the hypothesis that fibrils mediate neurotox-
icity [41]. Hence, the screening for compounds that reduce
oligomerization or formation of fibrils could be a discrete
target for therapeutics [42].
Replacement of the Pro with the Aze residue in iAb5p ana-
logue 1 causes an increase of the activity, suggesting that the
alteration of the backbone conformation associated with the en-
hancement of the ring constraint, when combined with reduc-
tion of the steric interference exerted by ring atoms [43],
improves destabilization of intermolecular b-sheet of Ab
aggregates. By taking into account this finding, the inactivity
of compound 2, containing the (2R,3R)-PhAze, doesn’t appear
in contrast with this result and underlines that the non-covalent
repulsive interactions exerted by ring atoms and by substituents
on the ring, in this crucial point of the molecule, exert an highly
unfavorable effect.
A residue of b-amino-carboxylic and of the corresponding
b-amino-ethanesulfonic acid, both possessing a pyrrolidine
ring, has been incorporated in compounds 3 and 4, respec-
tively. Although in both these analogues the inherent structural
flexibility of the b-residues is limited by the presence of the
pyrrolidine ring, this property, combined with the interference
exerted by the extra methylene group on the formation of
a regular hydrogen bonding network, results to be highly
unfavorable for the binding of the inhibitor to Ab [44].
Finally, compound 6 duplicates the iAb5 sequence and
maintains both the C-terminal amide and the N-terminal
acetyl protecting group. The inhibitory activity of this N-ace-
tyl-decapeptide-amide is very high and indicates an improved
interaction with the aggregating fibrils. This behaviour can be
related to that shown by the peptide dimers (bivalent ligands)
and their simultaneous interaction with different points of the
receptor [45]. These ligands, when can interact with two
proximal recognition sites of the receptor exhibit, due to
entropic factors, a considerably greater potency and affinity
than the monomeric analogues. In the present case the pic-
ture is more complex and strictly related to the interaction
mechanism of the dimer with the Ab molecule. Studies
focused on the activity of further new models, designed to
contain two monomeric fragments linked or cross-linked by
spacers of different length, seem now promising tools to
better understand this interesting point.
In summary, the here reported results confirm the value of
the peptide fragment individuated by Soto’s group as lead
compound to generate analogues capable to hinder fibrillogen-
esis. In particular we find that the enhancement of the

186
C. Giordano et al. / European Journal of Medicinal Chemistry 44 (2009) 179e189
conformational constraint at position 2 of the iAb5p, induced
by the incorporation of the Aze residue (analogue 1), improves
the activity. This effect, however, does not result compatible
with substituents on the Aze ring (analogue 2), thus indicating
the key role of the structural modifications at position 2 for
tuning the activity of the iAb5p analogues. Analogously, the
high activity of the peptide derivative 5 underlines the useful-
ness of modifications performed at the N-terminal position and
in particular the role that the sulfonamido junction can play in
terms of both metabolic stability and alteration of usual H-
bond patterns [44]. Finally, the dimeric analogue 6 may repre-
sent, as discussed above, the basis for the design of new
models to which several of the here reported results can be
transferred.
4.3. Coupling with carbodiimide
To an ice-cooled mixture containing the required N-pro-
tected amino acid or peptide (1.0 mmol), the C-protected
amino acid or peptide salt (1.0 mmol), HOBt (1.2 mmol) and
TEA (1.2 mmol) in anhydrous CH2Cl2 (6 mL), EDC
(1.0 mmol) were added and the reaction mixture was allowed
to warm slowly to room temperature overnight. The reaction
mixture was then diluted with CH2Cl2 (20 mL) and washed
with 1 M KHSO4 (2 ꢀ 15 mL), saturated aqueous NaHCO3
(2 ꢀ 15 mL) and brine (15 mL). The organic phase was dried
and evaporated under reduced pressure.
4.3.1. N-Benzyl-(2R,3R)-PhAze-PheeOMe
Purified on silica gel flash chromatography (hexane/CHCl3
7:3). Colourless oil (56%). IR n: 3687, 3356, 3029, 1743,
4. Experimental protocols
1
1668, 1514, 1236 cmꢃ1. H NMR d: 3.00e3.18 (m, 3H, Phe
CH2 and PhCH), 3.48 (d, 1H, J ¼ 10.0 Hz, Aze aCH),
3.60e3.82 [m, 7H, Aze CH2, PhCH2 and OCH3 (s at 3.76)],
4.90 (m, 1H, Phe aCH), 7.08e7.40 (m, 15H, Ar), 7.81 (d,
1H, J ¼ 6.0 Hz, NH). Anal. Calcd for C27H28N2O3: C,
75.68; H, 6.59; N, 6.54. Found: C, 75.73; H, 6.57; N, 6.49.
Melting points were determined with a Buchi B 540 appa-
¨
ratus and are uncorrected. IR spectra were recorded in 1%
CHCl3 (unless otherwise specified) solution employing a Per-
kineElmer FT-IR Spectrum 1000 spectrometer. 1H NMR
spectra were determined in CDCl3 solution with a Bruker
AM 400 spectrometer and chemical shifts were indirectly re-
ferred to TMS. The mass spectra were obtained in electrospray
(ES) conditions by a ZABSpec oa-TOF (Micromass Ltd.,
Manchester, UK) instrument. Thin-layer and preparative layer
chromatographies were performed on silica gel Merck 60 F254
plates. The drying agent was sodium sulfate. Elemental anal-
yses for C, H and N (where necessary a sample was further pu-
rified by preparative TLC) were performed in the laboratories
of the Servizio Microanalisi del CNR, Area della Ricerca di
Roma, Montelibretti, Italy, and were within ꢄ0.4% of the
theoretical values.
4.3.2. Cbzeb-HPro-PheeOMe
Purified on silica gel flash chromatography (EtOAc/Et2O).
Colourless oil (65%). IR n: 3686, 3420, 3010, 1741, 1685,
1
1514, 1418, 1226 cmꢃ1. H NMR d: 1.82e2.23 (m, 4H, Pro
CH2eCH2), 2.88e3.22 (two m, 4H, Phe CH2 and Pro
CH2N), 3.20 (br s, 2H, CH2CO), 3.64 (s, 3H, OCH3), 4.22e
4.38 (m, 1H, Pro aCH), 4.88e4.95 (m, 1H, Phe aCH), 5.16
(s, 2H, Cbz CH2), 6.48 (br s, 1H, NH), 7.00e7.41 (m, 10H,
Ar). Anal. Calcd for C24H28N2O5: C, 67.91; H, 6.65; N,
6.60. Found: C, 67.85; H, 6.72; N, 6.53.
4.3.3. CbzePhe-Asp(OtBu)eNH2
White solid (80%). IR n: 3685, 3402, 3034, 1724, 1498,
4.1. CbzeProej[CH2SO2]eCl
1
1156 cmꢃ1. H NMR (DMSO-d6) d: 1.40 [s, 9H, (CH3)3C],
Purified by silica gel flash chromatography (CH2Cl2). The
product was dried (P2O5) and stored under argon. Colourless
oil (46%). IR n: 3033, 2957, 2873, 1701 cmꢃ1. H NMR d:
1.92e2.12 (m, 4H, Pro CH2CH2), 3.18e3.42 (m, 2H, Pro
CH2N), 3.62e3.76 (m, 2H, CH2S), 4.61e4.75 (m, 1H, Pro
aCH), 5.18 (s, 2H, Cbz CH2), 7.38 (s, 5H, Ar). Anal. Calcd
for C13H16NO4ClS: C, 49.13; H, 5.08; N, 4.41. Found: C,
49.28; H, 4.97; N, 4.33.
2.40e2.82 [m, 3H, Phe CH2 (1H) and Asp CH2], 2.87e3.14
[m, 1H, Phe CH2 (1H)], 4.22 (m, 1H, Phe aCH), 4.50 (m,
1H, Asp aCH), 4.95 (s, 2H, Cbz CH2), 7.02e7.38 (m, 10H,
Ar), 7.55 (d, 1H, J ¼ 7.5 Hz, PheeNH), 8.22 (d, 1H,
J ¼ 7.5 Hz, AspeNH). Anal. Calcd for C25H31N3O6: C,
63.95; H, 6.65; N, 8.95. Found: C, 63.77; H, 6.59; N, 8.82.
1
4.3.4. N-Benzyl-(2R,3R)-PhAze-Phe-Phe-Asp(OtBu)eNH2
Crystallized from EtOAc. White solid (86%). IR n: 3658,
.
3401, 3034, 1674, 1497, 1200 cmꢃ1 1H NMR d: 1.39 [s,
4.2. CbzeProej[CH2SO2]ePheeOMe
9H, (CH3)3C], 2.58e3.15 (m, 5H, 2 ꢀ Phe CH2 and PhCH),
3.85e4.03 (m, 7H, Aze CH2, PhCH2, Asp CH2 and Aze
aCH), 4.10e4.64 (three m, 3H, 2 ꢀ Phe aCH and Asp
aCH), 6.98e7.51 (m, 20H, Ar). Anal. Calcd for
C43H49N5O6: C, 70.57; H, 6.77; N, 9.57. Found: C, 70.64;
H, 6.69; N, 9.71.
Purified by silica gel flash chromatography (CHCl3). Col-
ourless oil (48%). IR n: 3354, 3011, 1743, 1693, 1497,
1436, 1218 cmꢃ1 1H NMR d: 1.53e2.18 (m, 4H, Pro
.
CH2CH2), 2.45e3.24 (three m, 6H, Phe CH2, CH2S and Pro
CH2N), 3.78 (s, 3H, OCH3), 3.96e4.05 (m, 1H, Pro aCH),
4.30e4.43 (m, 1H, Phe aCH), 4.68e4.82 (m, 1H, SO2NH),
5.18 (s, 2H, Cbz CH2), 7.08e7.42 (m, 10H, Ar). Anal. Calcd
for C23H28N2O6S: C, 59.98; H, 6.13; N, 6.08. Found: C, 60.06;
H, 6.24; N, 6.17.
4.3.5. CbzeAze-Phe-Phe-Asp(OtBu)eNH2
Triturated with hexane. White solid (86%). IR n: 3676,
1
3400, 3022, 1685, 1517, 1234, 1156 cmꢃ1. H NMR d: 1.41

C. Giordano et al. / European Journal of Medicinal Chemistry 44 (2009) 179e189
187
[s, 9H, (CH3)3C], 2.19e2.43 (m, 2H, CH2CH2CH), 2.56e3.21
(m, 6H, 2 ꢀ Phe CH2 and Asp CH2), 3.65e3.80 and 3.84e
4.01 (two m, 2H, CH2CH2CH), 4.45e4.60 and 4.71e4.82
(two m, 4H, 4 ꢀ aCH), 5.10 (s, 2H, Cbz CH2), 5.21 (br s,
1H, NH), 6.39 (br s, 1H, NH), 6.70 (d, 1H, J ¼ 8.0 Hz, NH),
7.20e7.41 (m, 15H, Ar). Anal. Calcd for C38H45N5O8: C,
65.22; H, 6.48; N, 10.01. Found: C, 65.34; H, 6.57; N, 9.88.
4.3.10. AceLeu-Aze-Phe-Phe-Asp(OtBu)eNH2
Triturated with hexane. White solid (62%). IR n: 3326,
3009, 1673, 1513, 1369, 1235, 1156 cmꢃ1 1H NMR d:
.
0.89e0.98 (m, 6H, Leu CH3), 1.19e1.75 [m, 12H, Leu
CH2CH and (CH3)3C (s at 1.40)], 1.98 (s, 3H, CH3CO),
2.22e2.43 (m, 2H, CH2CH2CH), 2.52e3.15 (m, 6H,
2 ꢀ Phe CH2 and Asp CH2), 3.80e3.96 (m, 2H, CH2CH2CH
Aze CH2), 4.21e4.80 (m, 5H, 5 ꢀ aCH), 5.40 [s, 1H,
CONH2 (1H)], 6.08 (d, 1H, J ¼ 8.4 Hz, Leu NH), 6.42 [s,
1H, CONH2 (1H)], 6.70 (d, 1H, J ¼ 8.0 Hz, Phe NH), 6.98
(d, 1H, J ¼ 7.8 Hz, Phe NH), 7.04e7.31 (m, 10H, Ar), 7.82
(d, 1H, J ¼ 7.5 Hz, Asp NH). Anal. Calcd for C38H52N6O8:
C, 63.32; H, 7.27; N, 11.66. Found: C, 63.30; H, 7.22; N,
11.75.
4.3.6. Cbzeb-HPro-Phe-Phe-Asp(OtBu)eNH2
Crystallized from EtOAc/hexane. White solid (77%). IR n:
1
3692, 3402, 3036, 1681, 1498, 1236, 1157 cmꢃ1. H NMR d:
1.40 [s, 9H, (CH3)3C], 1.64e2.18 (m, 4H, Pro CH2CH2),
2.83e3.20 (m, 8H, 2 ꢀ Phe CH2, Pro CH2N and Asp CH2),
3.22e3.51 (m, 2H, Pro CH2CO), 4.15e4.98 (four m, 4H,
4 ꢀ aCH), 5.16 (s, 2H, Cbz CH2), 5.40, 6.37 and 6.66 (three
br s, 3H, 3 ꢀ NH), 7.05e7.42 (m, 15H, Ar). Anal. Calcd for
C40H49N5O8: C, 66.01; H, 6.79; N, 9.62. Found: C, 66.18;
H, 6.65; N, 9.48.
4.3.11. AceLeu-(2R,3R)-PhAze-Phe-Phe-Asp(OtBu)eNH2
Triturated with hexane. White solid (58%). IR n: 3331,
1
3017, 1673, 1498, 1236, 1156 cmꢃ1. H NMR d: 0.92e1.03
(m, 6H, Leu CH3), 1.22e1.73 [m, 12H, Leu CH2CH and
(CH3)3C (s at 1.40)], 2.00 (s, 3H, CH3CO), 2.63e3.26 (m,
7H, 2 ꢀ Phe CH2, PhCH and Asp CH2), 3.63e3.76 (m, 2H,
Aze CH2), 4.18e4.26 and 4.41e4.80 (two m, 5H, 5 ꢀ aCH),
5.50 [br s, 1H, CONH2 (1H)], 6.18 (br s, 1H, Leu NH), 6.37
[s, 1H, CONH2 (1H)], 6.64 (br s, 1H, Phe NH), 7.04e7.56
(m, 17H, Ar, Phe NH and Asp NH). Anal. Calcd for
C44H56N6O8: C, 66.31; H, 7.08; N, 10.55. Found: C, 66.17;
H, 7.23; N, 10.48.
4.3.7. CbzeProej[CH2SO2]ePhe-Phe-Asp(OtBu)eNH2
Pale yellow oil (55%). IR n: 3689, 3400, 3014, 1689, 1498,
1199 cmꢃ1. 1H NMR d: 1.43 [s, 9H, (CH3)3C], 1.72e2.08 (m,
4H, Pro CH2CH2), 2.41e3.25 (m, 10H, 2 ꢀ Phe CH2, Pro
CH2N, Asp CH2 and CH2SO2), 4.43e4.82 (m, 4H,
4 ꢀ aCH), 5.15 (s, 2H, Cbz CH2), 5.62 and 5.73 (two br s,
2H, 2 ꢀ NH), 6.31 (d, 1H, J ¼ 8.2 Hz, NH), 7.09e7.38 (m,
15H, Ar). Anal. Calcd for C39H49N5O9S: C, 61.32; H, 6.47;
N, 9.17. Found: C, 61.17; H, 6.62; N, 9.05.
4.3.12. AceLeu-b-HPro-Phe-Phe-Asp(OtBu)eNH2
Triturated with hexane. White solid (67%). IR n: 3690,
1
3411, 2930, 1709, 1498, 1369, 1222, 1156 cmꢃ1. H NMR
4.3.8. CbzeTau-Leu-Pro-PheeOMe
White solid (54%). IR n: 3675, 3409, 3032, 1668, 1512,
d: 0.86e1.08 (m, 6H, Leu CH3), 1.17e1.78 [m, 12H, Leu
CH2CH and (CH3)3C (s at 1.42)], 1.80e2.12 [m, 7H, Pro
CH2CH2 and CH3CO (s at 1.97)], 2.66e2.88 (m, 2H, Asp
CH2), 2.89e3.25 (m, 4H, 2 ꢀ Phe CH2), 3.28e3.46 (m, 2H,
Pro CH2CO), 3.60e3.98 (m, 2H, Pro CH2N), 4.20e4.95
(five m, 5H, 5 ꢀ aCH), 5.43 [s, 1H, CONH2 (1H)], 5.80 (br
s, 1H, Leu NH), 6.63 [s, 1H, CONH2 (1H)], 6.78 (br s, 1H,
Phe NH), 7.00e7.43 (m, 11H, Ar and Phe NH), 7.65 (br s,
1H, Asp NH). Anal. Calcd for C40H56N6O8: C, 64.15; H,
7.54; N, 11.22. Found: C, 64.10; H, 7.63; N, 11.03.
1
1356, 1212, 1150 cmꢃ1. H NMR d: 1.02e1.30 (m, 6H, Leu
CH3), 1.42e1.61 (m, 3H, Leu CH2CH), 1.70e2.19 (m, 4H,
Pro CH2CH2), 2.20e2.37 (m, 2H, Tau aCH2), 2.87e3.08
(m, 2H, Pro CH2N), 3.10e3.41 (m, 2H, Tau bCH2), 3.60e
3.82 (m, 2H, Phe CH2), 3.90 (s, 3H, OCH3), 4.12e4.82 (three
m, 3H, 3 ꢀ aCH), 5.36 (s, 2H, Cbz CH2), 5.80, 5.98 (two m,
2H, 2 ꢀ NH), 7.18 (d, 1H, J ¼ 6.0 Hz, NH), 7.30e7.65 (m,
10H, Ar). Anal. Calcd for C31H42N4O8S: C, 59.03; H, 6.71;
N, 8.88. Found: C, 59.19; H, 6.57; N, 8.71.
4.3.9. CbzeTau-Leu-Pro-Phe-Phe-Asp(OtBu)eNH2
4.3.13. AceLeu-Proej[CH2SO2]ePhe-Phe-
Asp(OtBu)eNH2
Triturated with hexane. White solid (67%). IR n: 3689,
Purified on silica gel flash chromatography (CHCl3/MeOH
98:2). White solid (40%). IR n: 3675, 3398, 3015, 1670, 1521,
1
1
1398, 1212, 1148 cmꢃ1. H NMR d: 1.12e1.22 (m, 6H, Leu
3401, 3027, 1673, 1509, 1396, 1202, 1153 cmꢃ1. H NMR
CH3), 1.52e1.65 [m, 12H, Leu CH2CH and C(CH3)3 (s at
1.62)], 1.98e2.38 (m, 6H, Pro CH2CH2 and Asp CH2), 3.00
(d, 2H, J ¼ 7.2, Tau aCH2), 3.14e3.58 (m, 6H, Pro CH2N
and 2 ꢀ Phe CH2), 3.60e3.78 (m, 1H, aCH), 3.80e3.98 (m,
2H, Tau bCH2), 4.28e5.19 (four m, 4H, 4 ꢀ aCH), 5.37 (s,
2H, Cbz CH2), 5.38 (d, 1H, J ¼ 6.1 Hz, NH), 5.93 (m, 1H,
NH), 5.83e6.04 [m, 2H, CONH2 (1H) and NH], 6.28 (d,
1H, J ¼ 6.3 Hz, NH), 6.70e6.82 [m, 1H, CONH2 (1H)],
7.20e7.62 (m, 16H, Ar and NH). Anal. Calcd for
C47H63N7O11S: C, 60.43; H, 6.80; N, 10.50. Found: C,
60.28; H, 6.75; N, 10.23.
d: 0.87e1.05 (m, 6H, Leu CH3), 1.29e1.74 [m, 12H, Leu
CH2CH and (CH3)3C (s at 1.46)], 1.86e2.18 [m, 7H, Pro
CH2CH2 and CH3CO (s at 2.02)], 2.88e3.24 (m, 8H,
2 ꢀ Phe CH2, Asp CH2 and CH2SO2), 3.48e3.67 (m, 2H,
Pro CH2N), 4.10e4.26, 4.62e4.68, 4.70e4.79 and 4.96e
5.04 (four m, 5H, 5 ꢀ aCH), 5.32 [s, 1H, CONH2 (1H)],
5.50 (br s, 1H, Leu NH), 6.10 (d, 1H, J ¼ 6.9 Hz, Phe NH),
6.32 [br s, 1H, CONH2 (1H)], 7.03 (d, 1H, J ¼ 6.5 Hz, Phe
NH), 7.10e7.32 (m, 10H, Ar), 7.42 (d, 1H, J ¼ 6.8 Hz, Asp
NH). Anal. Calcd for C39H56N6O9S: C, 59.67; H, 7.19; N,
10.71. Found: C, 59.52; H, 7.24; N, 10.63.

188
C. Giordano et al. / European Journal of Medicinal Chemistry 44 (2009) 179e189
4.4. N,N-Dimethyl-Tau-Leu-Pro-Phe-Phe-
Asp(OtBu)eNH2
4.6. Scanning electron microscopy studies
Morphologies of Ab1e40 were investigated using a scanning
electron microscopy (SEM) LEO 1450VP with tungsten fila-
ment as electron emitter with 20 keV acceleration potential
and a resolution of 4 nm. For the SEM study, samples of
Ab1e40 incubated in the absence and presence of inhibitors in
the same conditions as those described for amyloid detection
in suspension (see Section 4.5) were withdrawn at different
times, frozen, lyophilised and gold coated.
Triturated with hexane. White solid (quantitative). IR n:
1
3667, 3421, 3023, 1667, 1500, 1391, 1220, 1143 cmꢃ1. H
NMR (DMSO-d6) d: 0.73e0.95 (m, 6H, Leu CH3), 1.58e
1.75 [m, 12H, Leu CH2CH and C(CH3)3 (s at 1.60)], 2.38e
2.52 [m, 12H, Pro CH2CH2, Asp CH2 and (CH3)2N (s at
2.43)], 2.54 (d, 2H, J ¼ 6.80 Hz, Tau aCH2), 2.62e3.50 (m,
8H, Pro CH2N, 2 ꢀ PheeCH2 and Tau bCH2), 3.98e4.54
(five m, 5H, 5 ꢀ aCH), 6.81e7.27 [m, 11H, Ar and CONH2
(1H)], 7.83, 8.04, 8.18 (three m, 3H, 3 ꢀ NH), 8.32 [s, 1H,
CONH2 (1H)], 8.40 (m, 1H, NH). Anal. Calcd for
C41H61N7O9S: C, 59.47; H, 7.43; N, 11.84. Found: C, 59.62;
H, 7.38; N, 11.75.
4.7. Proteolytic assays
Aliquots of iAb5p and analogues at a concentration of
0.76 mg/mL in phosphate buffer saline (PBS) were incubated
over increasing time with 0.1 mg/mL of chymotrypsin (Fluka)
at 25 ꢁC. At the desired time, aliquots were withdrawn and
added to the same volume of the column elution mixture to
stop the reaction. This sample was immediately injected on
an RP-HPLC column (Vydac C18, 0.46 ꢀ 25 cm) and eluted
with an isocratic system (acetonitrile/water 80:20). The area
of the peak (UV absorbance at 215 nm) corresponding to the
intact peptide was measured and compared with an equivalent
sample incubated in PBS.
4.5. Amyloid detection in suspension
Ab1e40 (lots MZ09356, MZ13377 and MZ13736) was pur-
chased from NeoMPS (Strasbourg, France). The pretreatment
of Ab1e40 using the dissolution protocol in NaOH [22] was
used because this pretreatment was shown to facilitate amyloid
fibril formation. Typically, a 1 mg batch of Ab1e40 (around
2 ꢀ 10ꢃ4 mol with six acid equivalents) was added with no
more than 400 mL of 0.1 M NaOH (4 ꢀ 10ꢃ5 equiv OHꢃ),
sonicated in a water bath at room temperature and then added
with 1100 mL of MilliQ water to reach a final concentration of
0.67 mg/mL; the pH (Merck neutralyt) was around 7.5 and the
peptide was completely solubilized. The amount of NaOH
could be less than 4 ꢀ 10ꢃ5 equiv depending on the batch.
This sample was divided into several fractions, each contain-
ing about 100 mg of Ab1e40, frozen and lyophilised. Each
compound (1e6 and iAb5p) was dissolved with the minimum
amount of NaOH to titrate the solution to around pH 8.0
(Merck, neutralyt), buffered with 0.1 M Tris/HCl at pH 7.4
and then stored frozen at 3e4 mM. The incubation mixture
for amyloid fibril formation was prepared by dissolving
each lyophilised Ab1e40 fraction (about 100 mg) with
100 mL of 30 mM Tris/HCl buffer at pH 7.4 at 20 ꢁC with
and without 10 M excess of compound (1e6 and iAb5p) to
be tested as an inhibitor. The amyloid fibril formation was
tested according to LeVine [28] with minor modifications.
To detect and quantify the formation of amyloid fibrils, a 5
and 10 mL aliquots of the incubation mixture, in the presence
and absence of the inhibitor, were withdrawn at increasing
incubation times (0.5, 1.5, 6.0, 8.0, 15 and 24 h), and added
separately to two cuvettes containing 1 mM ThT each in
50 mM Tris/HCl buffer, pH 8.0 at 20 ꢁC. All experiments
were repeated three times and each sample was analyzed
in triplicates. Fluorescence emission spectra (460e550 nm)
were measured at 20 ꢁC in a PerkineElmer LS50B fluorim-
eter with excitation set at 450 nm immediately after addition
of the incubation mixture. The plots of the relative fluores-
cence at 485 nm as a function of the incubation times
(data not shown) indicate that in all the experiments the
relative fluorescence observed remained constant after 15 h
incubation.
4.8. Spectroscopy
Compounds 1e6 and iAb5p were dissolved as described in
Section 4.5, lyophilized and stored at ꢃ20 ꢁC before spectral
measurements. Far-UV CD (180e250 nm) measurements
were performed at 20 ꢁC in 0.01e0.1 cm path length quartz
cuvettes on each lyophilized compound dissolved in ultra
high quality water. CD spectra were recorded on a Jasco J-
720 spectropolarimeter. The results are expressed as the
mean residue ellipticity [Q] assuming a mean residue weight
of 110 per amino acid residue. FT-IR spectra in the amide I0
and amide II0 regions were recorded on a Nicolet Magna
760 spectrometer (Thermo Nicolet, Madison, WI, USA)
equipped with a liquid nitrogen-cooled mercuryecadmium
telluride solid-state detector. FT-IR transmission spectra of
2
about 20 mg/mL peptides in H2O were measured in a CaF2
cell with a 0.015 mm spacer (Spectra Tech, Madison, WI,
USA) at 20 ꢁC. For exchange of amide protons with deuterons,
each lyophilized compound (1e6 and iAb5p) was dissolved in
2H2O, freeze-dried and dissolved three times in 2H2O at 20 ꢁC
before measurements. A total of 512 interferograms at 2 cmꢃ1
resolution were collected for each spectrum, with Mertz apod-
ization and two levels of zero filling. The sample chamber of
the spectrometer was continuously purged with dry air to
avoid water vapor interference on the bands of interest. The
background spectra were collected immediately before the
sample measurements and under the same conditions with
the cell filled with everything but peptide sample. Water vapor
spectra were collected by reduction of the dry-air purge of the
clean cell. The analysis of raw spectra was performed with
GRAMS (Galactic Industries Corp., Salem, NH). The water
vapor contribution was subtracted from each of the sample

C. Giordano et al. / European Journal of Medicinal Chemistry 44 (2009) 179e189
189
[17] C. Soto, E.M. Sigurdsson, L. Morelli, R.A. Kumar, E.M. Castano,
B. Frangione, Nat. Med. 4 (1998) 822e826.
spectra. The essentially featureless region between 1700 and
1800 cmꢃ1 (where no peptide bands are present) indicated
that water vapor components are not responsible for the
observed bands in the amide I0 region. The individual compo-
nents in the amide region were identified by second derivative
and Fourier self-deconvolution of the raw spectrum using
GRAMS. Curve fitting of the raw spectrum with a mixed
GaussianeLorentzian function was then performed to deter-
mine the peak positions and parameters of each individual
component. The assignments of the component bands to
secondary structure elements were based on the literature
[37e39], and the area under each peak was used to calculate
the percentage of secondary structure components (Table 2).
[18] S.J. Wood, R. Wetzel, J.D. Martin, M.R. Hurle, Biochemistry 34 (1995)
724e730.
[19] F. Couty, G. Evano, M. Vargas-Sanchez, G. Bouzas, J. Org. Chem. 70
(2005) 9028e9031.
[20] S. Abele, K. Vogali, D. Seebach, Helv. Chim. Acta 82 (1999) 1539e1558.
[21] R.M.J. Liskamp, J.A.W. Kruijtzer, Mol. Divers. 8 (2004) 79e87.
[22] C. Giordano, G. Lucente, A. Mollica, M. Nalli, G. Pagani Zecchini,
M. Paglialunga Paradisi, E. Gavuzzo, F. Mazza, S. Spisani, J. Pept.
Sci. 10 (2004) 510e523.
[23] S.F. Nelsen, G.R. Weisman, Tetrahedron Lett. 26 (1973) 2321e2324.
[24] F. Couty, G. Evano, N. Ribasso, Tetrahedron: Asymmetry 14 (2003)
2407e2412.
[25] D. Braghiroli, R. Avallone, M. Di Bella, Tetrahedron: Asymmetry 8
(1997) 2209e2213.
[26] A.J. Brouwer, C.F. Monnee, R.M.J. Liskamp, Synthesis (2000) 1579e1584.
[27] E. Atherton, R.C. Sheppard, Solid Phase Synthesis e A Practical Ap-
proach, IRL Press, Oxford, 1989.
References
[28] H. LeVine, Protein Sci. 2 (1993) 404e410.
[29] Y. Fezoui, D.M. Hartley, J.D. Harper, R. Khurana, D.M. Walsh,
M.M. Condron, D.J. Selkoe, P.T. Lansbury Jr., A.L. Fink, D.B. Teplow,
Amyloid 7 (2000) 166e178.
[30] C. Goldsbury, P. Frey, V. Olivieri, U. Aebi, S.A. Muller, J. Mol. Biol. 352
¨
(2005) 282e298.
[1] L.C. Serpell, Biochim. Biophys. Acta 1502 (2000) 16e30.
[2] C. Adessi, C. Soto, Drug Dev. Res. 56 (2002) 184e193.
[3] C. Soto, M.E. Castano, B. Frangione, N.C. Inestrosa, J. Biol. Chem. 270
(1995) 3063e3067.
[4] J. Ghanta, C.L. Shen, L.L. Kiessling, R.M. Murphy, J. Biol. Chem. 271
(1996) 29525e29528.
[5] C. Soto, M.S. Kindi, M. Baumann, B. Frangione, Biochem. Biophys.
Res. Commun. 226 (1996) 672e680.
[31] R.A. Moore, S.F. Hayes, E.R. Fischer, S.A. Priola, Biochemistry 46
(2007) 7079e7087.
[32] K. Ma, L. Kann, K. Wang, Biochemistry 40 (2001) 3427e3438.
[33] A. Perczel, M. Hollosi, in: G.D. Fasman (Ed.), Circular Dichroism and
the Conformational Analysis of Biomolecules, Plenum Press, New
York, 1996, pp. 285e380.
[6] D. Gordon, S.C. Meredith, Biochemistry 42 (2003) 475e485.
[7] R.J. Chalifour, R.W. McLaughlin, L. Lavoie, C. Morissette, N. Tremblay,
´
´
M. Boule, P. Sarazin, D. Stea, D. Lacombe, P. Tremblay, F. Gervais, J.
[34] R.W. Woody, A.K. Dunker, in: G.D. Fasman (Ed.), Circular Dichroism
and the Conformational Analysis of Biomolecules, Plenum Press, New
York, 1996, pp. 109e157.
[35] B. Bochicchio, A. Pepe, A.M. Tamburro, Chirality 17 (2005) 364e372.
[36] S. Vuilleumier, J. Sancho, R. Loewenthal, A.R. Fersht, Biochemistry 32
(1993) 10303e10313.
Biol. Chem. 278 (2003) 34874e34881.
[8] F. Yang, G.P. Lim, A.N. Begum, O.J. Ubeda, M.R. Simmons,
S.S. Ambegaokar, P.P. Chen, R. Kayed, C.G. Glabe, S.A. Frautschy,
G.M. Cole, J. Biol. Chem. 280 (2005) 5892e5901.
[9] T. Cohen, A. Frydman-Marom, M. Rechter, E. Gazit, Biochemistry 45
(2006) 4727e4735.
[10] N. Kokkoni, K. Stott, H. Amijee, J.M. Mason, A.J. Doig, Biochemistry
45 (2006) 9906e9918.
[37] M. Jackson, H.H. Mantsch, Crit. Rev. Biochem. Mol. Biol. 30 (1995)
95e120.
[38] A. Barth, Prog. Biophys. Mol. Biol. 74 (2000) 141e173.
[39] A. Barth, C.Q. Zscherp, Rev. Biophys. 35 (2002) 369e430.
[40] Y. Pilpel, O. Bogin, V. Brumfeld, Z. Reich, Biochemistry 42 (2003)
3519e3526.
[11] T. Sato, P. Kienlen-Campard, M. Ahmed, W. Liu, H. Li, J.I. Elliott,
S. Aimoto, S.N. Constantinescu, J.-N. Octave, S.O. Smith, Biochemistry
45 (2006) 5503e5516.
[12] M. Torok, M. Abid, S.C. Mhadgut, B. Torok, Biochemistry 45 (2006)
5377e5383.
`
´
[41] C. Riviere, T. Richard, L. Quentin, S. Krisa, J.M. Merillon, J.P. Monti,
Bioorg. Med. Chem. 15 (2007) 1160e1167.
[42] D.J. Selkoe, Nature 399 (1999) A23eA31.
[13] C. Soto, Mol. Med. Today 5 (1999) 343e350.
[14] C. Hilbich, B. Kister-Woike, J. Reed, C.L. Masters, K. Beyreuther, J.
Mol. Biol. 228 (1992) 460e473.
[43] A. Zagari, G. Nemethy, H.A. Scheraga, Biopolymers 30 (1990) 951e
959.
[15] C. Adessi, M.J. Frossard, C. Boissard, S. Fraga, S. Bieler, T. Ruckle,
F. Vilbois, S.M. Robinson, M. Mutters, W.A. Banks, C. Soto, J. Biol.
Chem. 278 (2003) 13905e13911.
[44] R.C. Elgersma, T. Meijneke, R. de Jong, A.J. Brouwer, G. Postuma,
T.S. Dirk, D.T. Rijkers, R.M. Liskamp, J. Org. Biomol. Chem. 4
(2006) 3587e3597.
[16] B. Permanne, C. Adessi, G.P. Saborio, S. Fraga, M.J. Frossard, J. Van
Dorpe, I. Dewachter, W.A. Banks, F. Van Leuven, C. Soto, FASEB J.
16 (2002) 860e862.
[45] D. Cussac, M. Vidal, C. Leprince, W. Liu, F. Cornille, G. Tiraboschi,
B.P. Roques, C. Garbay, FASEB J. 13 (1999) 31e38.